• Inference Serving Stack: Scheduling and routing across heterogeneous models — the production path for vLLM and llm-d workloads on Kubernetes.
• Kubernetes-Native Workload Plumbing: Batch queueing, multi-cluster scheduling, and GPU sharing for AI/ML workloads — contributing upstream to Kueue, Karmada, and HAMi.
• Agent Harnesses: The supervision layer for long-running LLM agents — parallel sessions, multi-agent teams, artifacts, approval gates, scheduled dispatch. Because the work shouldn't collapse into chat.
• Agentic Developer Workflows: Multi-worktree dispatch, Claude Code skills, and commit/PR/task automation — the inner loop I live in daily.

• Mosoo (Core Contributor): Open-source, Cloudflare-native agent runtime for Codex, Claude Agent SDK, and OpenCode. stars
• OpenClaw : Open-source agent infrastructure for long-running, multi-channel AI work. stars
• semantic-router : Defining the decision-making layer for multi-model LLM serving. stars
• llm-d: Cloud-native infrastructure for disaggregated LLM inference. stars
• HAMi & Kueue: Kubernetes-native batch scheduling and GPU virtualization. HAMi stars Kueue stars
• Istio: Traffic governance for the service mesh layer. stars
• Karmada & Kubernetes: Multi-cluster orchestration foundations. Karmada stars Kubernetes stars

I build tools to fix my own problems.

• ClawWork: Client for OpenClaw — connect ClawWork to your own OpenClaw and unlock 10x multi-session productivity. release stars
• lathe: Agent-friendly CLI generator for APIs: turn Swagger, OpenAPI, and google.api.http protos into single-binary Cobra CLIs with catalogs and generated Skills. release stars
• kitup: Shared installer SDK for bundled Agent Skills — ship skills in your CLI, install them into Claude Code, Codex, and other agent hosts. release stars
• adit: AI-era notes you speak into being — macOS shell to talk with ChatGPT and Grok, then keep what matters in your local Library. mac stars
• awire: Cloudflare Worker control plane — route source events (GitHub) to delivery channels (Feishu) as interactive cards. Pure KV, HTMX admin, no build step. release stars
• agent-brains: Portable domain agent brains for coding agents — AGENTS.md, SOUL.md, MEMORY.md, skills, workflows, and a bundled loader. stars
• Recall: Local-first search across 10+ AI coding tools (Claude Code, Codex, OpenCode, Cursor, Gemini, Cline, Pi, Kiro, Copilot CLI, Antigravity) — hybrid FTS + embeddings, usage dashboard, all on-device. release stars
• gmc: Parallel git worktrees for parallel AI agents, plus AI-generated commits. release stars
• merge-scout: A Claude Code skill for vibe coding — ranks GitHub issues by contributability × merge probability so your agent picks work that will actually land. stars
• codex-agents-local: Local AGENTS.local.md overlays for Codex via hooks, without replacing the official codex command. stars
• pared: Minimal native macOS video compressor — drop a file, pick a style, done. mac stars
• mailbell: Minimal macOS menu bar notifier for Gmail. mac release stars
• mote: Menu bar macOS app for rewriting selected text with OpenAI-compatible models. mac stars
• branchlight: Checking PRs in browser tabs is a ritual. Moved it to the menubar. mac release stars
• MacMusicPlayer: A minimalist, clean music player for macOS. mac release stars
• SaveEye: A minimalist eye care reminder that doesn't annoy you. mac release stars
• Chrome TabBoost: Browser tab overload is a bug. I patched it with an extension. release stars
• ConfigForge: Manage ~/.ssh/config and kubeconfig without the headache. mac release stars
• LogoWallpaper: Generating brand assets shouldn't take 30 minutes. mac release stars
• prooflet: The proof layer for prototypes — in-page narration and annotation SDK, no backend in V1. stars
• mdctl: AI-powered Markdown workflow automation. release stars
• hf-model-downloader: Painless Hugging Face model downloads. release stars
• mirrormate: Docker pulls failing? I fixed it with magic. release stars
• swagger-online: Unified Swagger UI. No more tab chaos. release stars
• ai-icon-generator: I needed icons, so I built a generator. release stars
• gofs: python -m http.server is slow. Rewrote it in Go. release stars
• gh-x: Batch repo operations as a gh CLI extension. One-by-one doesn't scale. release stars
• convostore: Stateful LLM APIs shouldn't be this hard. A Redis-backed conversation store for OpenAI Responses API / vLLM. stars
• prompts: Prompt engineering is the real frontend of LLMs. I track mine here. release stars
• openclaw-gateway-tunnel (RFC, design phase): Zero-binary ngrok tunnel for OpenClaw gateways. Looking for co-designers before code lands. stars
• moltbot-channel-feishu (legacy, superseded by OpenClaw official): Feishu/Lark channel plugin for Moltbot/Clawdbot. release stars